facebookresearch/PyTorch-BigGraph

Health v2 · maintenance only

Flags: archived no_license

Adoption not part of the score

3454 stars · 454 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

PyTorch-BigGraph is a distributed system for learning embeddings of very large graph-structured data, scaling to billions of entities and trillions of edges. It supports multiple knowledge graph embedding models such as TransE, RESCAL, DistMult, and ComplEx, with optional multi-machine and GPU training.

Use cases

• generate embeddings for a large knowledge graph
• train node embeddings on billions of edges
• learn graph embeddings with TransE or ComplEx
• embed a large web interaction graph
• run distributed graph embedding training on multiple machines
• train graph embeddings on GPU

When to choose

• your graph is too large to fit in memory on a single machine
• you need scalable knowledge graph embedding training
• you want a PyTorch-based graph embedding framework from a reputable research lab

When to avoid

• your graph is small and fits in memory - simpler libraries like node2vec or PyTorch Geometric may suffice
• you need actively developed features - development activity is limited
• you need general-purpose GNN models rather than embedding-based link prediction

Facets

library · maturity maintenance

machine-learning deep-learning gpu-computing etl machine-learning large-language-models data-science python graph-embeddings knowledge-graphs distributed-training pytorch graph-partitioning algorithms linux gpu docker

Member repositories

For agents

markdown · JSON · MCP: product_card(name="facebookresearch/PyTorch-BigGraph")